Research

We study evolutionary biology, ecology, and cultural evolution using mathematical, computational, and statistical models and collaborations with experimental biologists.

Main interests

  • Generation & transmission of genetic and phenotypic variation
  • Changing & complex fitness landscapes
  • Microbial population biology
  • Cultural evolution

Research questions

  • How do different inheritance modes (high-fidelity vs. low-fidelity, vertical vs. horizontal) affect evolution?
  • How do these inheritance modes evolve and co-evolve?
  • How do socially-transmitted traits spread?
  • What are the similarities and differences between cultural and genetic evolution?
  • How do populations shift fitness peaks?
  • How do populations rapidly adapt to continuously changing environments?
  • How do ecology and evolution shape each other in the microbial world?
  • What are the microbial equivalents for concepts such as generation, age, death, birth, and most importantly, fitness?
  • What can new insights and models from machine and deep learning teach us about evolution?

Future projects

If you want to be involved, please email Yoav.

  • Co-evolution of learning and teaching
  • Transmission of connections in social networks with applications to animals and human social networks
  • Implementing evolutionary simulations using GPU and computational graph frameworks like TensorFlow
  • Similarities and differences between evolution and deep learning
  • Adaptive peak shifts with non-vertical inheritance
  • Modeling the role of aneuploidy in adaptive evolution
  • Using experimental data to predicting microbial competition results
  • Analysing results of evolutionary experiments using Approximate Bayesian Computation
  • Evolution of learning using neural networks as a cognitive model
  • Food culture: mining and surveying data on food culture, the set of practices, norms, and institution that determine what, when, and why we eat

Past research

Stress-induced mutation

We developed a theoretical basis to explain the evolution of stress-induced mutagenesis – the phenomena in which stress induces a transient increase in mutation rates. Stress-induced mutagenesis is prevalent in bacteria and empirical evidence suggests that it is common in many eukaryote species, from yeast to human cancer cells. We used mathematical models and computer simulations to show that (i) stress-induced mutagenesis is favored by natural selection (Ram & Hadany 2012; Ram, Altenberg, Liberman & Feldman, 2018); (ii) that this is also true in the presence of rare recombination (Ram & Hadany, in review); (iii) that stress-induced mutagenesis increases the rate of complex adaptation without reducing the mean fitness of the population (Ram & Hadany, 2014); (iv) errors in regulation of mutagenesis can be compensated by cell-to-cell signalling (Dellus-Gur et al., 2017).

Microbial evolution

We collaborated to develop and test a new method for predicting microbial growth in a mixed culture solely from growth curve data (Ram et al., submitted). To validate this method, we performed growth curve and competition experiments with bacteria. Our new method not only results in a simple and cost-effective approach for estimating growth in a mixed culture and inferring competitive fitness in microbes, but also provides information on the specific growth traits that contribute to differences in fitness, thus helping to bridge the gap between microbial ecology and evolution.

We also collaborated with the Kupiec lab at Tel Aviv University to analyse results of evolutionary experiments (Harari et al., 2018). In these experiments, haploid yeast cells frequently became diploid via two distinct genetic mechanisms. Using approximate Bayesian approximation (ABC), we estimated the rate of endoreduplication in these experiments, and found that in it was much higher than the mutation rate.

Cultural evolution

We studied the evolution of oblique transmission, in which offspring inherit their phenotype from a non-parental adult rather than their parents. This occurs, for example, in social learning, symbiont and pathogen transmission, and transfer of mobile genetic elements in microbes. We found that the evolutionarily stable rate of oblique transmission differs markedly from the rate that maximizes the geometric mean fitness of the population, implying a form of "prisoner's dilemma" for social learning (Ram, Liberman & Feldman, 2018).